Programming Languages and Compilers Commons^™

Boosting Chart-To-Code Generation In Mllm Via Dual Preference-Guided Refinement, Zhihan Zhang, Yixin Cao, Lizi Liao

Research Collection School Of Computing and Information Systems

Translating chart images into executable plotting scripts-referred to as the chart-to-code generation task-requires Multimodal Large Language Models (MLLMs) to perform fine-grained visual parsing, precise code synthesis, and robust cross-modal reasoning. However, this task is inherently under-constrained: multiple valid code implementations can produce the same visual chart, and evaluation must consider both code correctness and visual fidelity across diverse dimensions. This makes it difficult to learn accurate and generalizable mappings through standard supervised fine-tuning. To address these challenges, we propose a dual preference-guided refinement framework that combines a feedback-driven, dual-modality reward mechanism with iterative preference learning. Our approach introduces a structured …

Contrastrepair: Enhancing Conversation-Based Automated Program Repair Via Contrastive Test Case Pairs, Jiaolong Kong, Xiaofei Xie, Mingfei Cheng, Shangqing Liu, Xiaoning Du, Qi Guo

Research Collection School Of Computing and Information Systems

Automated Program Repair (APR) aims to automatically generate patches for rectifying software bugs. Recentstrides in Large Language Models (LLM), such as ChatGPT, have yielded encouraging outcomes in APR,especially within the conversation-driven APR framework. Nevertheless, the efficacy of conversation-drivenAPR is contingent on the quality of the feedback information. In this article, we propose ContrastRepair, anovel conversation-based APR approach that augments conversation-driven APR by providing LLMs withcontrastive test pairs. A test pair consists of a failing test and a passing test, which offer contrastive feedback tothe LLM. Our key insight is to minimize the difference between the generated passing test and the …

Auxiliary Prompt Tuning Of Vision‑Language Models For Few‑Shot Out‑Of‑Distribution Detection, Wenjun Miao, Guansong Pang, Zihan Wang, Jin Zheng, Xiao Bai

Research Collection School Of Computing and Information Systems

Recent advancements in CLIP-based out-of-distribution (OOD) detection have shown promising results via regularization on prompt tuning, leveraging background features extracted from a few in-distribution (ID) samples as proxies for OOD features.However, these methods suffer from an inherent limitation: a lack of diversity in the extracted OOD features from the few-shot ID data.To address this issue, we propose to leverage external datasets as auxiliary outlier data (i.e., pseudo OOD samples) to extract rich, diverse OOD features, with the features from not only background regions but also foreground object regions, thereby supporting more discriminative prompt tuning for OOD detection. We further introduce …

Probabilistic Prototype Calibration Of Vision-Language Models For Generalized Few-Shot Semantic Segmentation, Jie Liu, Jiayi Shen, Pan Zhou, Jan-Jakob Sonke, Stratis Gavves

Research Collection School Of Computing and Information Systems

Generalized Few-Shot Semantic Segmentation (GFSS) aims to extend a segmentation model to novel classes with only a few annotated examples while maintaining performance on base classes. Recently, pretrained vision-language models (VLMs) such as CLIP have been leveraged in GFSS to improve generalization on novel classes through multi-modal prototypes learning. However, existing prototype-based methods are inherently deterministic, limiting the adaptability of learned prototypes to diverse samples, particularly for novel classes with scarce annotations. To address this, we propose FewCLIP, a probabilistic prototype calibration framework over multi-modal prototypes from the pretrained CLIP, thus providing more adaptive prototype learning for GFSS. Specifically, FewCLIP …

Harnessing Se Community Knowledge For Developer-Centric Code Intelligence, Chengran Yang

Dissertations and Theses Collection (Open Access)

The integration of Large Language Models (LLMs), particularly those tailored for programming tasks—referred to as code LLMs—has created novel opportunities to enhance developer productivity. These advanced models automate routine and repetitive coding tasks, such as code generation and debugging, and enable faster prototyping and more efficient problem-solving. Despite these remarkable advantages, the current generation of code LLMs exhibits notable limitations that impact their practical effectiveness in real-world software engineering scenarios. These models frequently produce code that is inefficient or suboptimal in runtime performance, demonstrate opaque reasoning processes, and struggle to adapt effectively to diverse developer contexts and specific requirements. Moreover, …

Guiding Multiple Remote Users In Physical Tasks With Language-Driven Robotic Telepresence, Ruyi Li, Jingfei Guo, Xinyi Zhang, Xuji Zhang, Zeqing Li, Jiannan Li, Jiangtao Gong

Research Collection School Of Computing and Information Systems

Remote assistance through robotic telepresence could involve both control and memory challenges, particularly in one expert to multiple workers situation. In this work, we proposed a novelty language-driven interface to facilitate remote collaboration through telepresence robots. Through operations and maintenance expert interviews and a scenario simulation study, we identified key pain points in executing one-expert-multiple-workers remote guidance using the telepresence robot and proposed two design goals, which together consist of five sub-design goals with corresponding features. These features were integrated into a standard telepresence robot, resulting in the development of a Collaborative LLM-based Embodied Assistant Robot, named CLEAR Robot. A …

Exploring Parameter-Efficient Fine-Tuning Techniques For Code Generation With Large Language Models, Martin Weyssow, Xin Zhou, Kisub Kim, David Lo, Houari A. Sahraoui

Research Collection School Of Computing and Information Systems

Large language models (LLMs) demonstrate impressive capabilities to generate accurate code snippets given natural language intents in a zero-shot manner, i.e., without the need for specific fine-tuning. While prior studies have highlighted the advantages of fine-tuning LLMs, this process incurs high computational costs, making it impractical in resource-scarce environments, particularly for models with billions of parameters. To address these challenges, previous research explored in-context learning (ICL) and retrieval-augmented generation (RAG) as strategies to guide the LLM generative process with task-specific prompt examples. However, ICL and RAG introduce inconveniences, such as the need for designing contextually relevant prompts and the absence …

Detecting Defi Fraud With A Graph-Transformer Language Model, Wei Ma, Junjie Shi, Jiaxi Qiu, Cong Wu, Jing Chen, Lingxiao Jiang, Shangqing Liu, Yang Liu, Yang Xiang

Research Collection School Of Computing and Information Systems

With the rapid development of blockchain technology, the widespread adoption of smart contracts—particularly in decentralized finance (DeFi) applications—has introduced significant security challenges, such as reentrancy attacks, phishing, and Sybil attacks. To address these issues, we propose a novel model called TrxGNNBERT, which combines Graph Neural Network (GNN) and the Transformer architecture to effectively handle both graph-structured and textual data. This combination enhances the detection of suspicious transactions and accounts on blockchain platforms like Ethereum. TrxGNNBERT was pre-trained using a masked language model (MLM) on a dataset of 60,000 Ethereum transactions by randomly masking the attributes of nodes and edges, thereby …

Automatic Generation Of Introductory Programming Exercises With Large Language Models, Nguyen Binh Duong Ta, Hua Gia Phuc Nguyen, Gottipati Swapna

Research Collection School Of Computing and Information Systems

Despite recent advances in code generation made possible by large language models (LLMs), programming is still an essential skill that computing students need to master now and in the foreseeable future. In learning programming, frequent practices with exercises set at an appropriate difficulty and knowledge level is of crucial importance for students. However, it’s not a trivial task for instructors to create many good quality exercises customized for each student. Programming problems found on Internet sources such as LeetCode are mostly too challenging for novice programmers with no prior coding knowledge. Recent work in AI-enabled education has been leveraging LLMs …

Llm2rec: Large Language Models Are Powerful Embedding Models For Sequential Recommendation, Yingzhi He, Xiaohao Liu, An Zhang, Yunshan Ma, Tat‑Seng Chua

Research Collection School Of Computing and Information Systems

Sequential recommendation aims to predict users' future interactions by modeling collaborative filtering (CF) signals from historical behaviors of similar users or items. Traditional sequential recommenders predominantly rely on ID-based embeddings, which capture CF signals through high-order co-occurrence patterns. However, these embeddings depend solely on past interactions, lacking transferable knowledge to generalize to unseen domains. Recent advances in large language models (LLMs) have motivated text-based recommendation approaches that derive item representations from textual descriptions. While these methods enhance generalization, they fail to encode CF signals-i.e., latent item correlations and preference patterns-crucial for effective recommendation. We argue that an ideal embedding model …

How To Enable Effective Cooperation Between Humans And Nlp Models: A Survey Of Principles, Formalizations, And Beyond, Chen Huang, Yang Deng, Wenqiang Lei, Jiancheng Lv, Tat-Seng Chua, Jimmy Huang

Research Collection School Of Computing and Information Systems

With the advancement of large language models (LLMs), intelligent models have evolved from mere tools to autonomous agents with their own goals and strategies for cooperating with humans. This evolution has birthed a novel paradigm in NLP, i.e., human-model cooperation, that has yielded remarkable progress in numerous NLP tasks in recent years. In this paper, we take the first step to present a thorough review of human-model cooperation, exploring its principles, formalizations, and open challenges. In particular, we introduce a new taxonomy that provides a unified perspective to summarize existing approaches. Also, we discuss potential frontier areas and their corresponding …

How Developers Use Type-System Related Programming Language Features, Samuel W. Flint

School of Computing: Dissertations, Theses, and Student Research

Optional type annotations are a popular feature of programming languages that allow developers to omit explicit type information in code while, in some cases, retaining many of the benefits of static typing, such as in-code documentation, improved detection of type errors, or enforcement of code properties. However, how developers use and understand optional type annotations is not clear. The focus of this dissertation is to understand the use and comprehension of optional type annotations.

Optional type annotations are examined through four lenses: first, by examining the evolution of usage in a statically typed programming language (Kotlin, the default language for …

How Developers Use Type-System Related Programming Language Features, Samuel W. Flint

Dissertations and Doctoral Documents, University of Nebraska-Lincoln, 2023–

Optional type annotations are a popular feature of programming languages that allow developers to omit explicit type information in code while, in some cases, retaining many of the benefits of static typing, such as in-code documentation, improved detection of type errors, or enforcement of code properties. However, how developers use and understand optional type annotations is not clear. The focus of this dissertation is to understand the use and comprehension of optional type annotations.

Optional type annotations are examined through four lenses: first, by examining the evolution of usage in a statically typed programming language (Kotlin, the default language for …

Equivalence And Similarity Refutation For Probabilistic Programs, Krishnendu Chatterjee, Ehsan Kafshdar Goharshady, Petr Novotný, Dorde Zikelic

Research Collection School Of Computing and Information Systems

We consider the problems of statically refuting equivalence and similarity of output distributions defined by a pair of probabilistic programs. Equivalence and similarity are two fundamental relational properties of probabilistic programs that are essential for their correctness both in implementation and in compilation. In this work, we present a new method for static equivalence and similarity refutation. Our method refutes equivalence and similarity by computing a function over program outputs whose expected value with respect to the output distributions of two programs is different. The function is computed simultaneously with an upper expectation supermartingale and a lower expectation submartingale for …

Debate, Reflect, And Distill: Multi-Agent Feedback With Tree-Structured Preference Optimization For Efficient Language Model Enhancement, Xiaofeng Zhou, Heyan Huang, Lizi Liao

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) continue to set new standards in knowledge-intensive and complex reasoning tasks, yet their high computational demands limit widespread adoption. While distilling large models into smaller ones offers a sustainable solution, current techniques—such as static knowledge distillation, resource-intensive reinforcement learning from human feedback, or limited self-reflection—struggle to yield substantial and lasting performance gains. In this paper, we present a novel Debate and Reflect (D&R) framework that orchestrates multi-turn debates between smaller models and stronger teacher models, eliciting actionable feedback (e.g., error analysis, corrective strategies) to guide student models. Further, we introduce Tree-structured Direct Preference Optimization (T-DPO) to …

Memotune: A Measure And Moment-Driven Fine-Tuning Framework For Quantized Large Language Models, Yun Zhang, Xue Geng, Lizi Liao, Jintong Sun, Minghe Yu, Ge Yu

Research Collection School Of Computing and Information Systems

Quantizing large language models (LLMs) is essential for reducing memory and computational costs in natural language processing. Existing methods combine quantization with parameter-efficient fine-tuning but often fail to meet practical performance requirements. This paper introduces MeMoTune, a novel fine-tuning framework for quantized LLMs. By employing a measure and moment approach within a low-rank approximation framework in probability measure space, MeMoTune optimizes the objective function for superior fine-tuning results. The update process is further refined through scaled gradient, enhancing convergence efficiency and noise robustness. Experiments on tasks like text generation, summarization, and understanding show MeMoTune significantly outperforms state-of-the-art methods, e.g. fine-tuning …

Focus: Evaluating Pre-Trained Vision-Language Models On Underspecification Reasoning, Kankan Zhou, Yibin Lai, Kyriakos Mouratidis, Jing Jiang

Research Collection School Of Computing and Information Systems

Humans possess a remarkable ability to interpret underspecified ambiguous statements by inferring their meanings from contexts such as visual inputs. This ability, however, may not be as developed in recent pre-trained visionlanguage models (VLMs). In this paper, we introduce a novel probing dataset called FOCUS to evaluate whether state-of-the-art VLMs have this ability. FOCUS consists of underspecified sentences paired with image contexts and carefully designed probing questions. Our experiments reveal that VLMs still fall short in handling underspecification even when visual inputs that can help resolve the ambiguities are available. To further support research in underspecification, FOCUS will be released …

Evaluating And Mitigating Linguistic Discrimination In Large Language Models: Perspectives On Safety Equity And Knowledge Equity, Guoliang Dong, Haoyu Wang, Jun Sun, Xinyu Wang

Research Collection School Of Computing and Information Systems

By training on text in various languages, large language models (LLMs) typically possess multilingual support and demonstrate remarkable capabilities in solving tasks described in different languages. However, LLMs can exhibit linguistic discrimination due to the uneven distribution of training data across languages. That is, LLMs are hard to keep the consistency of responses when faced with the same task but depicted in different languages. In this study, we first explore the consistency in the LLMs’ outputs responding to queries in various languages from two aspects: safety and quality. We conduct this analysis with two datasets (AdvBench and NQ) based on …

Evowiki: Evaluating Llms On Evolving Knowledge, Wei Tang, Yixin Cao, Yang Deng, Jiahao Ying, Bo Wang, Yizhe Yang, Yuyue Zhao, Qi Zhang, Xuanjing Huang, Yu-Gang Jiang, Yong Liao

Research Collection School Of Computing and Information Systems

Knowledge utilization is a critical aspect of LLMs, and understanding how they adapt to evolving knowledge is essential for their effective deployment. However, existing benchmarks are predominantly static, failing to capture the evolving nature of LLMs and knowledge, leading to inaccuracies and vulnerabilities such as contamination. In this paper, we introduce EvoWiki, an evolving dataset designed to reflect knowledge evolution by categorizing information into stable, evolved, and uncharted states. EvoWiki is fully auto-updated, enabling precise evaluation of continuously changing knowledge and newly released LLMs. Through experiments with Retrieval-Augmented Generation (RAG) and Continual Learning (CL), we evaluate how effectively LLMs adapt …

Fact-Audit: An Adaptive Multi-Agent Framework For Dynamic Fact-Checking Evaluation Of Large Language Models, Hongzhan Lin, Yang Deng, Yuxuan Gu, Wenxuan Zhang, Jing Ma, See-Kiong Ng, Tat-Seng Chua

Research Collection School Of Computing and Information Systems

Large Language Models (LLMs) have significantly advanced the fact-checking studies. However, existing automated fact-checking evaluation methods rely on static datasets and classification metrics, which fail to automatically evaluate the justification production and uncover the nuanced limitations of LLMs in fact-checking. In this work, we introduce FACT-AUDIT, an agent-driven framework that adaptively and dynamically assesses LLMs’ fact-checking capabilities. Leveraging importance sampling principles and multi-agent collaboration, FACT-AUDIT generates adaptive and scalable datasets, performs iterative model-centric evaluations, and updates assessments based on model-specific responses. By incorporating justification production alongside verdict prediction, this framework provides a comprehensive and evolving audit of LLMs’ factual reasoning …